Treatment-Aware Hyperbolic Representation Learning for Causal Effect Estimation with Social Networks

1 code implementation • 12 Jan 2024 • Ziqiang Cui, Xing Tang, Yang Qiao, Bowei He, Liang Chen, Xiuqiang He, Chen Ma

Firstly, TAHyper employs the hyperbolic space to encode the social networks, thereby effectively reducing the distortion of confounder representation caused by Euclidean embeddings.

Representation Learning